Ranolazine inhibits NaV1.5-mediated breast cancer cell invasiveness and lung colonization

Abstract

Background: Na(V)1.5 voltage-gated sodium channels are abnormally expressed in breast tumours and their expression level is associated with metastatic occurrence and patients' death. In breast cancer cells, Na(V)1.5 activity promotes the proteolytic degradation of the extracellular matrix and enhances cell invasiveness.

Findings: In this study, we showed that the extinction of Na(V)1.5 expression in human breast cancer cells almost completely abrogated lung colonisation in immunodepressed mice (NMRI nude). Furthermore, we demonstrated that ranolazine (50 μM) inhibited Na(V)1.5 currents in breast cancer cells and reduced Na(V)1.5-related cancer cell invasiveness in vitro. In vivo, the injection of ranolazine (50 mg/kg/day) significantly reduced lung colonisation by Na(V)1.5-expressing human breast cancer cells.

Conclusions: Taken together, our results demonstrate the importance of Na(V)1.5 in the metastatic colonisation of organs by breast cancer cells and indicate that small molecules interfering with Na(V) activity, such as ranolazine, may represent powerful pharmacological tools to inhibit metastatic development and improve cancer treatments.

Figure 1

Ranolazine inhibits sodium current in human breast cancer cells. Sodium currents (INa) from MDA-MB-231 breast cancer cells stably expressing null target shRNA (shCTL) were studied in voltage-clamp mode with the whole-cell configuration of the patch clamp technique. A, Left, representative INa-voltage traces obtained from one cell before (vehicle) and after 50 μM ranolazine treatment (Rano). Right, mean ± s.e.m. steady-state INa-voltage relationships obtained from cancer cells before and after incubation with 50 μM ranolazine (n = 12 cells) from a holding potential of −100 mV. There is as statistical difference between the two conditions for voltages ranging from −35 to +40 mV (p < 0.001, Wilcoxon test). B, Availability-voltage relationships obtained in presence (red trace) or not (vehicle, black trace) of 50 μM ranolazine. There is a significant leftward shift of the availability-voltage relationship in presence of ranolazine (p < 0.001). The half (1/2)-inactivation voltage was shifted from −84.1 ± 1.4 mV to −90.3 ± 1.7 mV in absence and presence of ranolazine, respectively. C, Activation-voltage relationships obtained in presence (red trace) or not (vehicle, black trace) of 50 μM ranolazine. There is a significant leftward shift of the activation-voltage relationship in presence of ranolazine, and the 1/2-activation voltage was shifted from −37.1 ± 1.0 mV to −39.2 ± 0.6 mV in absence and presence of ranolazine, respectively. (p < 0.01, Wilcoxon test).

Figure 2

Ranolazine inhibits the Na _V 1.5-mediated breast cancer cell invasiveness in vitro. A, SCN5A mRNA expression assessed by real-time qPCR in shCTL and shNa_V1.5 cells (n = 10 separate experiments) and compared with a Mann–Whitney test. B, Mean ± s.e.m. peak INa recorded in 23 shCTL cells and in 20 shNa_V1.5 cells under a depolarization from −100 to −5 mV (Mann–Whitney test). Representative currents are shown underneath. C, shCTL and shNa_V1.5 cell growth and viability after 5 days, expressed relative to the shCTL cell line (n = 3 independent experiments). D, Cell viability of shCTL after 5 days of growth in presence of increasing concentrations of ranolazine, from 0.1 to 100 μM, and expressed relative to the control condition without ranolazine (vehicle). E, Effect of 30 μM tetrodotoxin (TTX) or 50 μM ranolazine (Rano) on shCTL and shNa_V1.5 human breast cancer cell invasiveness (Kruskal-wallis analysis followed by a Dunn’s test). F, shCTL and shNa_V1.5 cells were cultured for 24 h on a Matrigel™-composed matrix treated with 50 μM ranolazine (Rano) or not. F-actin cytoskeleton was stained with phalloidin-AlexaFluor594. A cell circularity index was calculated using ImageJ© software (n = 138–238 cells analysed, Mann–Whitney test). G, shCTL cells were cultured on a Matrigel™-composed matrix containing DQ-Gelatin® for 24 h in presence or not of 50 μM ranolazine. A “Matrix-Focalized-degradation activity index” was calculated as being the number of pixels corresponding to the co-localization of F-actin condensation areas (F-actin cytoskeleton was stained with phalloidin-Alexa594) and focal spots of DQ-gelatin proteolysis (coloc) (7). Results are expressed relative to the control condition (CTL, N = 534 cells) without ranolazine (Rano, N = 375 cells) and compared using Mann–Whitney test. Representative pictures are shown on the left. Statistical significance is indicated as: *p <0.05; **p < 0.01 and ***p < 0.001. NS stands for not statistically different.

Figure 3

Na _V 1.5 suppression, or ranolazine treatment, inhibit metastatic lung colonisation by breast cancer cells. A, Representative bioluminescent imaging (BLI) measurement performed in the same NMRI nude mouse per condition from week 2 to week 8 after cancer cell injection. Mice were injected with shCTL MDA-MB-231-Luc cells (shCTL), or with shNa_V1.5 MDA-MB-231-Luc cells (shNa_V1.5) or with shCTL MDA-MB-231-Luc cells and treated (5 days/week) with ranolazine (50 mg/kg) (Rano) or vehicle (shCTL, shNa_V1.5). B, Evolution of mice body weight during the experiments in the same conditions than in A. C, Mean in vivo BLI value (expressed in cpm) as a function of time recorded in the whole body of mice coming from the three groups indicated previously (shCTL, n = 18; shNa_V1.5, n = 12; Rano, n = 8) (Statistical significance is indicated as: *p <0.05, Kruskal-Wallis analysis followed by Dunn’s test). D, Representative BLI at completion of the study (8^th week after cells injection), in whole animals and ex vivo after lung isolation. E, BLI quantification of excised lungs. Box plots indicate the first quartile, the median, and the third quartile, squares indicate the mean (shCTL, n = 18; shNa_V1.5, n = 12; Rano, n = 8) (Kruskal-Wallis analysis followed by Dunn’s test).